Suggestions for SPSS enhancements

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Suggestions for SPSS enhancements

Dennis Deck
I'm not sure where to send suggestions for future SPSS versions any
more.  
Here is a partial list.


DOCUMENTATION (syntax guide)
- Add the version number in which each command was added.  Also note any
updates to syntax in subsequent versions (at least the most recent).  
- Include more examples.  The old SPSS manuals used to provide many more
examples than current manuals.  For syntactically or conceptually
complex procedures this was particularly helpful.  


MATCH FILES
Bring back the local /Keep and /Drop (or add /LKeep and /LDrop to stay
backward compatible to recent versions).
In the original syntax, /Keep and /Drop were local to the preceeding
/File or /Table subcommand.    
Some versions ago this was changed to apply globally to the result file.

In my experience it is almost always harder to use this way.  


MIXED  
Add link functions so it can handle binary and count outcome variables.

In many fields, binary outcomes are the norm rather than the exception.
I should think that this will be easier now with that GENLIN was added.



PROC SQL
Add the equivalent of SAS's PROC SQL
This would greatly enhance the data manipulation capabilities of SPSS.

This should be a little easier now with the addition of commands like
DATASET NAME.  


GET CAPTURE ODBC
This is particularly hard to use and trouble shoot, though this is
admittedly partly an ODBC problem.  
The wizard is helpful as a starting point but the generated syntax is
pretty unreadable.  Better formatting would be a start.  
It would also help to have better documentation, more diagnostic error
messages, and troubleshooting tips for this command.  
For example, when an error is encountered there is nothing to indicate
where the error occurred.  
While SPSS can't do much to change those error messages generated by the
ODBC driver, it could provide guidance for common errors.



Dennis Deck, PhD
RMC Research Corporation
111 SW Columbia Street, Suite 1200
Portland, Oregon 97201-5843
voice: 503-223-8248 x715
voice: 800-788-1887 x715
fax:  503-223-8248
[hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Suggestions for SPSS enhancements

zstatman
Dennis

Will log for Development

Tks,
WMB
Primary SPSS Beta Site


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Dennis Deck
Sent: Wednesday, December 27, 2006 3:31 PM
To: [hidden email]
Subject: Suggestions for SPSS enhancements

I'm not sure where to send suggestions for future SPSS versions any more.
Here is a partial list.


DOCUMENTATION (syntax guide)
- Add the version number in which each command was added.  Also note any
updates to syntax in subsequent versions (at least the most recent).
- Include more examples.  The old SPSS manuals used to provide many more
examples than current manuals.  For syntactically or conceptually complex
procedures this was particularly helpful.


MATCH FILES
Bring back the local /Keep and /Drop (or add /LKeep and /LDrop to stay
backward compatible to recent versions).
In the original syntax, /Keep and /Drop were local to the preceeding
/File or /Table subcommand.
Some versions ago this was changed to apply globally to the result file.

In my experience it is almost always harder to use this way.


MIXED
Add link functions so it can handle binary and count outcome variables.

In many fields, binary outcomes are the norm rather than the exception.
I should think that this will be easier now with that GENLIN was added.



PROC SQL
Add the equivalent of SAS's PROC SQL
This would greatly enhance the data manipulation capabilities of SPSS.

This should be a little easier now with the addition of commands like
DATASET NAME.


GET CAPTURE ODBC
This is particularly hard to use and trouble shoot, though this is
admittedly partly an ODBC problem.
The wizard is helpful as a starting point but the generated syntax is
pretty unreadable.  Better formatting would be a start.
It would also help to have better documentation, more diagnostic error
messages, and troubleshooting tips for this command.
For example, when an error is encountered there is nothing to indicate
where the error occurred.
While SPSS can't do much to change those error messages generated by the
ODBC driver, it could provide guidance for common errors.



Dennis Deck, PhD
RMC Research Corporation
111 SW Columbia Street, Suite 1200
Portland, Oregon 97201-5843
voice: 503-223-8248 x715
voice: 800-788-1887 x715
fax:  503-223-8248
[hidden email]
Will
Statistical Services
 
============
info.statman@earthlink.net
http://home.earthlink.net/~z_statman/
============